OBJECTIVE: In this study, we assessed the temporal trends and relative and attributable perinatal risks of maternal obesity over a 20-year period. STUDY DESIGN: We conducted a retrospective cohort study between 1980 and 1999 by using a computerized perinatal database of all women who received prenatal care and delivered their infants within a regional health care system The main outcome measures were as follows: (1) annual mean body weight and the percentage of women classified as obese at the first prenatal visit (primary definition ≥200 lb; secondary definitions ≥250 lb, ≥300 lb, body mass index >29 kg/m2); and (2) relative and attributable risks of obesity for selected maternal and perinatal morbidities in successive 5-year periods. RESULTS: From 1980 to 1999, the mean maternal weight of women at the first prenatal visit increased 20% (144-172 lb), as did the percentage of women ≥200 lb (7.3-24.4), the percentage ≥250 lb (1.9-10.7), the percentage ≥300 lb (0.5-4.9), and the percentage with a body mass index >29 kg/m2 (16.3-36.4), P<.01 for all. Controlling for maternal age, race, and smoking status, obese women were at increased risk at each period for cesarean delivery (range of adjusted relative risk, 1.5-1.8), gestational diabetes (range, 1.8-2.9), and large (>90th percentile) for gestational age infants (range, 1.8-2.2). From the earliest 5-year period (1980-1984) to the most recent (1995-1999), the percentage of obesity-attributable cesarean deliveries more than tripled from 3.9 to 11.6. Similar percentage increases were observed for the obesity-attributable risks for gestational diabetes (12.8-29.6) and large for gestational age infants (6.5-19.1). Trends for secondary obesity definitions were similar, although the magnitude of the increased attributable risks was smaller. CONCLUSIONS: Efforts to reduce the frequency of certain perinatal morbidities will be constrained unless effective measures to prevent, or limit the risks of, maternal obesity are developed and implemented.